Soil temperature is a critical parameter influencing ecological and hydrological processes, yet its accurate projection under climate change remains challenging due to coarse-resolution climate models and complex soil-atmosphere interactions. This study develops a deep learning framework to downscale soil temperature (5 cm depth) in western Iran, under climate change scenarios. Using an ensemble of three complementary techniques—Random Forest (Gini) importance, Permutation Importance, and SHAP (SHapley Additive exPlanations) analysis—we identified optimal predictors from the 26 available CanESM5 (CMIP6) variables. Four deep learning models—CNN, LSTM, GRU, and a hybrid CNN-LSTM—were evaluated for downscaling performance using historical data (1980–2014). The hybrid CNN-LSTM model outperformed others, achieving the highest accuracy (NSE > 86%, RMSE < 4.3°C) by capturing spatial and temporal dependencies in soil thermal dynamics. Assessing the plausibility of each scenario’s trend (2015–2020) revealed regional climate patterns: western stations, more arid and warming-sensitive, aligned with SSP245/SSP370, while eastern stations, influenced by the Zagros Mountains, showed cooling and precipitation feedbacks favoring SSP119/SSP126. Future projections under SSP126, SSP245, and SSP585 scenarios indicated nonlinear soil temperature responses, with high-emission pathways (SSP585) causing initial cooling (-4.11°C by 2040) followed by accelerated warming (+ 2.09°C by 2100). In stark contrast, low- and mid-emission pathways (SSP126, SSP245) lead to stable, moderate warming. This creates a dramatic reversal in decadal trends for SSP585, shifting from strong cooling to rapid heating, and alters the climate’s statistical profile. The findings emphasize that high emissions defer but ultimately cause the most intense warming, highlighting the critical influence of emission pathways on the future pace and pattern of soil temperature change.
Saeidinia et al. (Thu,) studied this question.